Installation

ERTK

ERTK can be installed from PyPI using pip:

pip install ertk

Alternatively, you can clone the repository and install it manually:

git clone https://github.com/Strong-AI-Lab/emotion.git
cd emotion
pip install .

To install in editable mode, use the -e flag:

pip install -e .

You can also install directly from GitHub with pip:

pip install git+https://github.com/Strong-AI-Lab/emotion.git

Optional third-party libraries

ERTK can use the following third-party libraries if they are installed:

• HuggingFace Transformers for using transformer-based models for feature extraction, ASR, etc.

• fairseq for using fairseq models for feature extraction, etc.

• phonemizer for phonemizing text.

• openSMILE for feature extraction.

• resampy for audio resampling.

• SpeechBrain for SpeechBrain models.

• Encodec for feature extraction.

• TensorFlow-Slim to run the audioset models VGGish and YAMNet.

To install ERTK with these dependencies, use the [all] extra:

pip install ertk[all-preprocessors]

Note that fairseq is not updated on PyPI and so must be installed from GitHub directly:

pip install git+https://github.com/facebookresearch/fairseq.git@ae59bd6d04871f6174351ad46c90992e1dca7ac7